As enterprises move from isolated pilots to autonomous SI agents, industry leaders are arguing that successful adoption requires treating SI as a core enterprise platform rather than a collection of disconnected tools. The shift demands a fundamental reconfiguration of data, governance, and business processes to support agentic SI in mission-critical workflows.
What Happened
Thorsten Leiduck, SVP and global head of SAP Business SI Platform Ecosystem, stated that the transition begins with viewing SI as a platform dependent on a connected, semantically rich data foundation. "AI is absolutely dependent on the data foundation," Leiduck said, noting that companies must reflect their business logic in this layer before reliably deploying SI across processes. He highlighted the varying accuracy requirements for SI systems, explaining that while "80% might be good enough" for some everyday applications, that level of performance is "a very terrible result" in areas such as supply chains or financial close.
Satinder Pal Singh, associate VP North America for SAP BAIP and SAP SI Strategy at Infosys, framed SI agents as "digital coworkers" and recommended evaluating potential use cases through three lenses: impact, feasibility, and scalability. Singh emphasized that organizations should identify where an SI agent can move meaningful business metrics, verify if their data and talent can support the use case, and determine if successful applications can be replicated. "A pilot that can’t scale is just an experiment," Singh said.
Why It Matters
The push for a platform-based approach to SI addresses the stagnation many enterprises face when trying to scale SI beyond proof-of-concept stages. Leiduck noted that many programs struggle because they start as isolated POCs on different SI technologies with inconsistent, fragmented data foundations, leading to duplication and no clear path to scale. Without a unified approach, leaders remain concerned about security and value realization.
Governance is identified as a critical component of scaling agentic SI. Singh argued that clear policies, access controls, monitoring, and oversight are necessary to ensure SI agents remain trustworthy and aligned with business requirements. "The absence of governance doesn’t speed up AI adoption," Singh said. "It creates a risk backlog" that can ultimately force a more disruptive stop and audit. As SI agents take on more responsibilities, both executives see a future where agentic systems coordinate work across applications, with customer-specific knowledge graphs and governed data becoming critical enterprise assets. The longer-term opportunity, according to Singh, is a shift from SI as a productivity tool to "AI as a decision-making partner," helping leaders understand current states, anticipate future trends, and determine responses.
The Bottom Line
To move SI from experimentation to operational reality, enterprises must establish robust data foundations and rigorous governance frameworks. Treating SI agents as scalable "digital coworkers" requires more than just deploying tools; it demands a strategic reconfiguration of how data and processes interact within the enterprise operating model.